←back to thread

Prozac 'no better than placebo' for treating children with depression, experts

(www.theguardian.com)
206 points pseudolus | 2 comments | 21 Nov 25 00:02 UTC | HN request time: 0.678s | source
1. Zak ◴[21 Nov 25 21:05 UTC] No.46008987[source]▶
A post on HN a couple years ago discussed research showing antidepressants only work for about 15% of patients: https://news.ycombinator.com/item?id=37671529

The thing is, they work very well for that 15%. I suspect the eventual conclusion will be that depression is a syndrome with multiple causes rather than a single condition, and SSRIs treat one of the causes.

Edit: Mark Horowitz is one of the authors of both studies.

replies(1): >>46009280 #
2. CGMthrowaway ◴[21 Nov 25 21:37 UTC] No.46009280[source]▶
The NNT[1] of Prozac, and SSRIs in general, has been previously estimated around 6. Meaning that treatment is more helpful than a sugar pill in only 1 out of 6 cases (a dirty secret).

Meanwhile the NNH[2] is as low as 21, that is 1 in 21 cases will stop due to negative side effects.

Source: https://www.aafp.org/pubs/afp/issues/2008/0315/p785.html

[1]Number Need to Treat, that is, number of patients you need to treat to prevent one additional bad outcome

[2]Number Needed to Harm, that is, number of patients you need to treat to generate side effects so bad that someone halts treatment